Slashdot Mirror
Intel Opens Its Front-Side Bus
vivin writes "The Inquirer is reporting that Intel has opened up its FSB. Intel did this during IDF 07. What this means is that you can plug non-Intel things into the Intel CPU socket. The article says 'This shows that Intel is willing to take AMD seriously as a competitive threat, and is prepared to act upon it. In addition to this breaking one of the most sacred taboos at Intel, it also hints that engineering now has the upper hand over bureaucracy.'"
This isn't the first time socket sharing has occured
The old Socket 7 used to fit Intel and AMD and Cyrix.
Hell, it can even house socket 5 cpus!
Back then it wasn't a big deal to upgrade a CPU.
All the companies started changing sockets at a frantic pace and made a simple CPU update essentially mean a whole machine.
A new motherboard for the new socket but it also has new memory footprint as well so that gets replaced, and the PCIx slot won't fit my agp card.
liqbase
I hope so. Every time I have to upgrade my machine I have to spend an hour on the web working out the 700 different kinds of processor I can buy and what type of socket I need to support them.
I had an AMD Duron 800MHz that I tried to replace with an Athlon 1300MHz which should have been supported, but created a nifty column of smoke when I plugged it in. Anything that reduces that likelihood is good in my book.
Peter
There are no AMD chips that you could plug into it. It is not that Intel created a socket/bus that can take AMD chips. The news is that they opened it so that their competitors can develop chips for their socket/bus if they would desire to do so. So in the future we may see AMD chips that will fit into Intel FSB, but I doubt that will happen in the near future.
>> 'This shows that Intel is willing to take AMD seriously as a competitive threat, and is prepared to act upon it.'
I'm not sure how much sense this statement really makes. If they take AMD as a serious threat, wouldn't they WANT AMD to be forced to continue using their own bus? AM2 was probably a misstep, given the performance drops, giving intel the upper hand, but now they are willing to let AMD play in their sandbox - it helps AMD more than it hurts them.
I'm not complaining about the move, I just found the article a bit sparse on details and the statement at odds with common sense. Is it fully open, or does it require licensing? What is AMD's take on this news? How much re-work will be required to move AMD's processor cores to the intel bus? Will they gain performance or lose it in the translation?
Lots of questions that the Inquirer seems to totally ignore in what may be a significant development in the battle of the big boys.
so why would anybody want to plug in an AMD processor there unless it was hugely cheaper or more powerful?
For starters, intel's frontside bus is just that, a good old-fashioned FSB that hasn't changed much in years.
AMD's processors have something completely different. Not only is it physically incompatible, it's actually "Hypertransport" which is marketing speak for a chip-to-chip interconnect. Look at all the big iron manufacturers supporting it. Note no intel. AMD has been shipping these processors since 2003. Intel's (incompatible) equivalent isn't due out until 2008. Other manufacturers have been shipping CPUs with similar interconnects since the mid 1990s (UltraSPARC, MIPS).
AMD processors implement NUMA via this interconnect. Each CPU can have its own local memory. On an intel system, all processors compete for bandwidth over the shared FSB
This is why Opteron/Athlon 64 systems scale well past 2 processors. This is also why it will be easier to make e.g. graphics processors that fit in AMD motherboards.
intel processors may currently do better on selected synthetic benchmarks and niche applications. AMD, however, has a far more sophisticated, modern and scalable platform. Intel set sail on the itanic.
I'm not complaining about the move, I just found the article a bit sparse on details and the statement at odds with common sense. Is it fully open, or does it require licensing? What is AMD's take on this news? How much re-work will be required to move AMD's processor cores to the intel bus? Will they gain performance or lose it in the translation?
Intel is not trying to open their bus up to AMD. That is not at all the goal. First of all, access to the the Intel bus requires a license. I'm not sure Intel would even grand AMD one for a sane price. Second of all, AMD would in no way want Intel's bus. As has been the hot topic of discussion for over a year, AMD's HT (HyperTransport) point-2-point links are faster both in terms of bandwidth, and latency than Intel's FSB. HT uses less pins than Intel's bus, and HT devices are simpler, cost less, and use less power. HT is a pretty neat and effective technology. Intel's FSB on the other hand, is much the same as it was around 10 years ago. To answer your question, AMD would take a massive hit by going to Intel's POS bus. It's funny, ATM, AMD has the better bus/platform and Intel has the better core. No one here seems to realize that AMD would never be willing to throw out their main advantage right now... AM2 isn't the issue. The issue is HT. Hell, even IBM announced that Power7 will use AMD's HT links. No one will be dropping HT for the POSFSB any time soon.
Intel/AMD are only opening their sockets/buses in an attempt to get third party developers to make FPGAs, JAVAics, and other accelerators. AMD has had some luck with this, and one can buy co-processors that drop into an AMD socket today. Intel is trying to get the same benifits, but I don't really see the point until Intel can get CSI working and drop the antiquated FSB.
I am a viral sig. Please help me spread.
AMD opened their HyperTransport bus, royalty free, in 2001. They've signed people like Sun and Cisco, who have a big interest in moving a lot of data on buses. And if you get people using your bus, you can easily talk them into using your processors in their embedded devices.
That was a while ago, but I suspect it's coming to fruition or perhaps gaining more traction, if only now Intel is saying "me too."
http://news.zdnet.com/2100-9595_22-528221.html
Give it ten years and what will we have? 256-core processors running with core clock speeds of 100GHz?
100GHz is probably pushing it. You'll note that we haven't seen a huge increase in clock speeds recently, but rather continuing increases in instructions per cycle. I'd guess we'll reach a plateau somewhere around the 10GHz mark.
Moore's Law will soon hit a much more fundamental law: physics. You can't keep shrinking transistors like they are at the moment; it was predicted that we'd reach the limit years ago (yes, I too remember the advent of 200MHz desktop processors, and thinking they couldn't get much faster), but the fact we haven't so far doesn't mean we won't. Moore's Law demands a shrinking by a factor of 1.4 every 18 months. We're currently on 45nm. This gives us the following trend:
end 2008 - 32nm
start 2010 - 22nm
end 2011 - 16nm
start 2013 - 12nm
end 2014 - 8nm
start 2016 - 6nm
end 2017 - 4nm
4 nanometres is only 38 atomic radii of silicon. It seems unlikely that a transistor this small could be produced. Therefore, as long as we continue to use silicon transistors (and no promising alternative that solves this issue exists right now) we will see the end of Moore's Law within the next 10 years. I'm sure of it.
And an end of Moore's Law will not only slow GHz increases, but also will slow the adoption of larger numbers of cores, because without shrinking transistors the only way to increase number of cores is by having a larger die size, which is more expensive and requires larger chip size, which requires larger system board size, which requires larger case size, which consumers don't like.